KR100886994B1 - How to estimate error rate at receiver using iterative decoding - Google Patents

Abstract

A method of estimating an error rate of a received communication includes repeatedly processing each of a plurality of received symbol sequences to provide each of a plurality of data sequences, and the number of iterations performed to provide each of the plurality of data sequences. Providing a step. The error rate of the received communication can then be estimated based on the number of iterations performed to provide each of the plurality of data sequences. Related receivers and communication devices are also discussed.

Radio Processor, Repetitive Baseband Processor, Error Detection Circuit, Error Estimator

Description

How to estimate error rate in receiver using iterative decoding {METHOD FOR ESTIMATING ERROR RATES IN RECEIVERS USING ITERATIVE DECODING}

TECHNICAL FIELD The present invention relates to the field of communications, and in particular, to a method for estimating error rate and an associated receiver.

In a digital communication system such as a wireless telephone communication system, the quality of service may be defined by the average frame error rate (FER) or average bit error rate (BER) for the received information frame. To maintain the desired quality of service, the transmitter (for a mobile terminal or base station) may, for example, adjust the transmit power to adjust the desired quality of service, changes in system loading (number of mobile terminals served), and / or changes in channel conditions. It is possible to change the level and / or channel / source coding method. Thus, an estimate of the error rate for the communication received at the receiving device can be calculated and provided to the transmitting device and used to change the transmit power.

The error rate can be estimated by averaging a number of bits or frames that fail cyclic redundancy check. However, this method can provide a relatively slow convergent speed. If the operating frame error rate is 1%, for example, a statistically reliable frame error rate may require hundreds of frame observations. During this period, the channel conditions can change significantly, so that reactive adaptation can be difficult to achieve.

According to an embodiment of the invention, the error rate of the received communication can be estimated. In particular, each of the plurality of received symbol sequences may be processed repeatedly to provide each of the plurality of data sequences, and the number of iterations performed to provide each of the plurality of data sequences may be provided. The error rate of the received communication can be estimated based on the number of iterations performed to provide each of a plurality of data sequences.

1 and 2 are block diagrams illustrating a receiver and method in accordance with the present invention.

3A-B and 4A-B are graphs illustrating the correlation between error rate and processor iteration in accordance with an embodiment of the invention.

5 and 6 are flowcharts illustrating the operation of a receiver and method according to an embodiment of the present invention.

7 shows an embodiment of a lookup table according to the invention.

FIG. 8 is a table illustrating operating conditions for the simulations of FIGS. 19 to 22.

9, 10, 17, 18, and 19 illustrate a simulation of a receiver and method according to the present invention.

11, 12, 13, 14, 15, 16, 20, 21, and 22 show a simulation of a conventional receiver in graphical form.

23 is a block illustrating a mobile terminal according to the present invention.

The invention will now be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments are shown. However, the invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Identical numbers are referred to by identical elements from beginning to end.

As will be appreciated by those skilled in the art, the present invention may be embodied in a method or apparatus. Thus, the present invention may appear as a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. When an element is referred to as being "connected" or "connected" to another element, it will also be understood that it may be directly connected or connected to another element, or an arbitration element may be provided. In contrast, when an element is referred to as being "directly connected" or "directly connected" to another element, no arbitration element is provided.

As shown in the block diagram of FIG. 1, a receiver in accordance with the present invention comprises an antenna 21, a radio processor 23, an iterative baseband processor 25, an error detection circuit 27, and an error estimator 29. It may include. The radio processor generates soft values at a symbol rate for the received communication, and the repeating baseband processor decodes the soft values using an iterative processing technique.

A second example of a receiver according to the invention is shown in FIG. 2, where the repeatable baseband processor 25 ′ is a first processor 41, a second processor 43, a deinterleaver 45, and an interleaver ( 47). Like the receiver of FIG. 1, the receiver of FIG. 2 also includes an antenna 21, a radio processor 23, an error detection circuit 27, and an error estimator 29. The baseband processor 25 ′ may be a turbo decoder, where the first and second processors 41 and 43 may be first and second map decoders, respectively. Optionally, the baseband processor 25 'may be a turbo equalizer, where the first processor 41 may be an equalizer and the second processor 43 may be a map decoder.

The repeatable baseband processor 25 is described, for example, in IEEE International Conference On Communications Conference ICC '93, May 23-26, 1993, pp. It may be a turbo decoder for parallel connected convolutional codes as discussed in "Near Shannon Limit Error-Correcting Coding And Decoding: Turbo-Codes" by Claude Berrou et al., 1064-1070, the disclosure of which is incorporated herein by reference in its entirety. have. The turbo decoding algorithm may use soft-input soft-output (SISO) to iteratively process component code. Other repeating baseband processors are described, for example, by a repeating decoder for serially connected code (IEEE Transactions On Information Theory, Vol. 44, No. 3, May 1998, pp. 909-926, by Sergio Benedetto, et al., "Serial Concatenation Of Interleaved Codes"). ; Performance Analysis, Design, and Iterative Decoding "); Decoding Algorithm for Iterative Modulation and Coding Difference Modulation (by Krishna R. Narayanan et al., IEEE Transaction On Communications, Vol. 47, No. 7, July 1999, pp. 956-961, "A Serial Concatenation Approach To Iterative Demodulation And Decoding "); And decoding algorithms for iterative equalization and coding multipath channels ("Turbo-Detection: A New Approach To Combat Channel Frequency Selectivity" by Proceeding Of IEEE International Communications Conference '97, 1997, pp. 1498-1502). Include. The disclosures for each of the aforementioned references are incorporated herein by reference in their entirety.

A common aspect of an iterative baseband processor is that the average operating complexity (and power consumption) can be significantly reduced without terminating performance by terminating iterative processing on received frames as the resulting data passes error detection / correction. Is there. The turbo decoder can be designed, for example, to perform eight iterations for a certain operating point. However, error-free decoding can only be achieved in multiple data frames after two or three iterations of the repeating baseband processor. When a frame is received with a significant amount of noise and the repeating baseband processor converges relatively slowly, additional iteration processing may only be required in rare cases.

Thus, the error detection circuit 27 can be used to detect the accuracy of the estimated data sequence generated by the repeating baseband processor 25, and the estimated data sequence can be used to detect error using an error detection technique such as cyclic redundancy check. When passing (or can be corrected) it can be used to stop subsequent iteration processing on the received data frame. In other words, the iterative processing for the symbol sequence can be stopped when a correction data sequence (eg, a data frame) can be provided. This can reduce the average operating complexity without reducing performance. The end of the iterative process for the symbol sequence upon detecting the correct sequence of estimation data is described, for example, in Proceeding Of IEEE International Symposium On Information Theory '98, 16-21 August 1998, pp. 279, Rose Y. Shao. "Two Simple Stopping Criteria For Iterative Decoding"), Reference 2 (US Patent No. 5,761,248 by Joachim Hagenauer, filed June 2, 1998, "Method And Arrangement For Determining An Adaptive Abort Criterion In Iterative Decoding) Of Multi-Dimensionally Coded Information "); And "Complexity Reduction Of Turbo Decoding" by Akira Shibutani et al., Proceeding Of IEEE Vehicular Technology Conference '00 Fall, October 1999, pp. 1570-1574. The disclosures for each of the aforementioned references are incorporated herein by reference in their entirety.

Iterative processing can be stopped using, for example, cyclic redundancy check (CRC) bits provided in data frames of many communication systems. In this embodiment, the CRC bits may be inserted into each data frame and the CRC check sum may be calculated at the end of each iteration. Thus, the CRC code can be used to be instructed when the iterative processor generates a corrected data sequence for the data frame, and this indication can be used to stop subsequent iterative processing for the data frame, thereby avoiding the repetitive operation performed by it. Potentially reduced.

Upon detecting the correction data sequence, the error detection circuit 27 provides repetitive control feedback to the repetitive baseband processor 25, which instructs the repetitive baseband processor to stop repetitive processing for the symbol sequence and detects the error. The circuit 27 passes the correction data sequence like the information bits used in the communication device. The error detection circuit also passes the number of iterations performed to generate a sequence of correction data to the error estimator 29, where the number of iterations performed can be used to calculate an estimate of the error rate for the received communication.

For example, the average number of iterative processes used to generate a predetermined number of consecutive correction data sequences can be used to estimate the error rate. In particular, the error estimator may rely on the finding that the average number of iterative processes is generally strongly correlated to the operating frame error rate (FER) and the bit error rate (BER). This correlation is shown in Figures 3A-B and 4A-B, and the receiver according to the invention comprises a wideband-CDMA (W-CDMA) turbo decoder.

In Figures 3A-B, the correlation between the operation bits and the frame error rate, and the average number of decoding iterations, refer to a W-CDMA receiver comprising a Log-Lin turbo decoder of the repeating baseband processor 25. Is shown. Examples of log-rin turbo decoders are described, for example, in Proceeding Of IEEE Vehicular Technology Conference '00 Spring, May 2000, pp. It is discussed with reference to "Linearly Approximated Log-Map Algorithms For Turbo Decoding" by Jung-Fu Cheng et al., 2252-2256, the disclosure of which is incorporated herein by reference in its entirety. In FIG. 3A, the operating frame error rate (FER) for a log-lean turbo decoder is decoded per data frame for two different channel models and three different data frame sizes (K) (K is the number of bits in the data frame). The average number of iterations is shown. Two other channel models are additive white Gaussian noise (AWGN) and independent Rayleigh fading (IRF) channels. In FIG. 3B, the operating bit error rate (BER) for a log-lean turbo decoder is shown as the average number of decoding iterations per data frame for two different channel models and three different frame sizes.

delete

In Figures 4A-B, the correlation between the operating error rate and the average number of decoding iteration processes is shown for a W-CDMA receiver that includes a Log-Max turbo decoder for the repeating baseband processor 25. have. Examples of log-maximum turbo decoders are described, for example, in Proceedings Of IEEE International Communications Conference '95, June 1995, pp. It is discussed with reference to "A Comparison Of Optimal And Sub-Optimal MAP Decoding Algorithms Operating In The Log Domain" by Patrick Robertson et al., 1009-1013, the disclosure of which is incorporated herein by reference in its entirety. In FIG. 4A, the operating frame error rate (FER) for the log-maximum turbo decoder is decoded per data frame for two different channel models and three different data frame sizes (K) (K is the number of bits in the data frame). The average number of iterations is shown. Two other channel models are additional white Gaussian noise (AWGN) and independent Rayleigh fading (IRF) channels. In Figure 4B, the operating bit error rate (BER) for the log-maximum turbo decoder is shown as the average number of decoding iterations per data frame for two different channel models and three different frame sizes.

Although the correlation of error rates with respect to the average number of decoding iterations has been discussed above with respect to receivers including log-lean and log-max turbo decoders, such iterative decoders have been discussed as examples, and the method and receiver according to the invention are different. It may be performed using an iterative baseband processor and / or an iterative baseband processing technique. Thus, for repetitive baseband processor performance and a given data frame size K, a reliable correlation between the average number of baseband repetition processing and the operation bit and / or frame error rate can be determined. In addition, FIGS. 3A-B and 4A-B show that the correlation between the operating error rate and the average number of decoding iterations is not significantly affected by the selection of the channel model.

An example operation for the receiver of FIG. 1 is shown in the flowchart of FIG. 5. As discussed above, the receiver repeatedly processes each successive received symbol sequence to provide each of a plurality of data sequences (eg, data frames) to block 51. For each of the provided data sequences, a number of repetitive processes performed to provide the data sequence are provided to block 53, and the number of repetitive processes performed is used to estimate the error rate of the communication received at block 55. For example, the average number of iterations performed can be calculated using the most recently recorded number of iterations, along with the number of previous iterations performed on the previous data sequence, the average of which is shown in FIGS. 3A-B and 4A-. It can be used to estimate the error rate using data such as that provided in B. The operation of FIG. 5 may be repeated for additional received signal sequences at decision block 57.                 

Another example of the receiver operation of FIG. 1 is shown in FIG. 6. In particular, iterative baseband processor 25 iteratively processes the received symbol sequence to provide the data sequence estimated at block 61 at the end of each iteration. The error detection circuit 27 performs an error check on the data sequence estimated at the end of each iteration process. If the data sequence fails error checking at decision block 63 and the maximum number of iterations are not performed for the received signal sequence at decision block 65, the repeating baseband processor 25 may then process the subsequent estimated data sequence. Subsequent iterations are performed at block 61 to provide. Thus, the iterative processing for the received signal sequence is iteratively controlled until the estimated data sequence passes error detection in the determination block 63 or until the maximum number of iterations are performed as indicated in the determination block 65. Iterate in response to feedback.

As the iteration process for the received symbol sequence (identified by the signal sequence m) is achieved, a number of iterations N _m performed to produce the final estimated data sequence m is written to block 67. do. In addition, the estimated data sequence m can be passed like an information bit (e.g., a data frame) as shown in FIG. When block 67 is reached, when the maximum number of iterations is performed without passing error detection, N _m may be the maximum number of iterations and the final estimated data sequence m passes like a data bit. Can be. Optionally, N _m may be ignored if the last estimated data sequence does not pass error detection, and / or if the last estimated data sequence may be discarded.

)를 갱신하도록 사용될 수 있다. 수행된 반복 처리의 평균 수는 예컨대 슬라이딩 윈도우 평균화(sliding window averaging) 또는 지수 포겟팅 평균화(exponential forgetting averaging)와 같은 평균화 기술을 사용하여 계산될 수 있다. 슬라이딩 윈도우 평균화를 사용할 때, 에러 추정기(29)는 반복 처리의 수(N_n, N_m-1... N_m-L+1)를 제공하도록 지난 리스트(L)를 유지하고, 반복 처리의 현재 평균은 다음과 같이 계산될 수 있다:

The number of iterations (N _m ) for the most recent estimate data sequence is then the average number of iterations performed at block 69 (

) Can be used to update. The average number of iterative processes performed can be calculated using averaging techniques such as sliding window averaging or exponential forgetting averaging. When using sliding window averaging, the error estimator 29 maintains the past list L to provide the number of iterations (N _n , N _m-1 ... N _{m-L + 1} ), The current average can be calculated as follows:

delete

)를 반복 처리의 현재 수(Nm)와 결합시킨다:

When exponential forgetting averaging is used, the error estimator 29 uses the exponential forgetting factor (0 <

) Is combined with the current number of iterations (Nm):

delete

Thus, when using exponential forgetting averaging, the past list that reported the number of iterations (N _n , N _m-1 ... N _{m-L + 1} ) is not required. Although sliding window averaging and exponential forgetting averaging have been discussed as examples, other averaging techniques can be used.

)를 사용하여 블록(71)에서 에러 충정기(29)에 의해 추정될 수 있다. 특히, 동작 FER 및/또는 BER은 예컨대 도 3A-B 및 도 4A-B에 그래픽 식으로 도시된 에러율에 관한 관계와 함께 수행된 반복 처리 평균 수를 사용하여 추정될 수 있다. 도 3A-B 및 도 4A-B에 그래프 식으로 도시된 관계는 룩업 테이블과 같이 또는 적절한 수학적 관계 또는 기능과 같이 에러 추정기(29)에 이용가능한 메모리에 저장될 수 있다. 수행된 반복 처리 평균 수 및 에러율 사이의 관계는 사용된 반복 기저대역 프로세서의 수행, 데이터 시퀀스(즉, 데이터 프레임 크기(K))내의 비트 수, 사용된 채널 모델, 및/또는 다른 요인에 종속될 수 있고, 다수의 관계(즉 룩업 테이블 또는 수학적 기능)는 다른 동작 조건을 조정하도록 제공될 수 있다. 게다가, 수행된 반복 처리 평균 수 및 에러율 사이의 관계는 시간 및/또는 갱신된 온라인에 걸쳐 트레인(trained)될 수 있다.The motion frame error rate (FER) and / or bit error rate (BER) is the average number of iterations performed at that time (

) Can be estimated by the error corrector 29 at block 71. In particular, the operation FER and / or BER can be estimated using, for example, the number of iterations performed, together with the relationship regarding the error rate graphically shown in FIGS. 3A-B and 4A-B. The relationships graphically depicted in FIGS. 3A-B and 4A-B may be stored in memory available to the error estimator 29 as a lookup table or as a suitable mathematical relationship or function. The relationship between the average number of iterations performed and the error rate may depend on the performance of the iterative baseband processor used, the number of bits in the data sequence (ie, data frame size K), the channel model used, and / or other factors. Multiple relationships (ie, lookup tables or mathematical functions) can be provided to adjust other operating conditions. In addition, the relationship between the average number of iterations performed and the error rate can be trained over time and / or updated online.

The estimation of the error rate can then be sent to the wireless device transmitting the received communication and can be used to adjust the transmit power for the communication received at the receiver of FIG. Increasing the error rate for communications received at the receiver of FIG. 1 may indicate the need to transmit communications from other wireless devices at high power to reduce the error rate for communications received at the receiver of FIG. Optionally, a reduction in the error rate for the communication received at the receiver of FIG. 1 may indicate that another wireless device may transmit at low power. The receiver of FIG. 1 may then proceed with the iterative processing of the subsequent received symbol sequence if there is a subsequent received symbol sequence as indicated in the determination block 73.

According to the present invention, multiple iterative baseband iterations performed for each received symbol sequence can be used to estimate the error rate for the received communication. Thus, the error rate estimate can be updated for each received frame, and the transmission loss characteristic of each update error estimate can be greater than one. That is, the number of repeat baseband iterations performed for each received symbol sequence may vary between 1 and the maximum number of allowed iterations. In addition, statistically reliable error rate estimation can be obtained by monitoring a relatively small number of processed frames.

In one example of a receiver according to the invention, the repeatable baseband processor 25 may be a log-max decoder having a data frame size of K = 640 and a maximum number of iterations equal to eight, which is 0.01 frame error rate for this description. It can be assumed to be currently operating with (FER). As shown in Figure 4A, the corresponding average number of decoding iterations for each data frame is nearly three. When using the error estimation method according to an embodiment of the present invention, monitoring as little as 20 data frames can be used to generate a reliable error rate estimate, and the number of decoding iterations is monitored every frame, thereby monitoring whether the frame is in error. can do. This implementation can be achieved where a measurement (i.e. number of iterations) is provided for each frame and each measurement has a transmission loss characteristic of 8 (maximum number of iterations). In contrast, conventional error estimators may rely on monitoring actual bits / frames and need to monitor the order of at least several hundred frames to obtain a reliable frame error rate at FER = 0.01.

Simulation results will be discussed with reference to FIGS. 9, 10, 17, 18, and 19 for a W-CDMA turbo code receiver in accordance with an embodiment of the present invention that receives 10,000 data frames, where iterative baseband. The processor is a log-maximum decoder using a data frame size of K = 640 and sliding window averaging. The relationship between the frame error rate for this W-CDMA receiver and the average number of decoding iterations is shown in FIG. 4A, and the curve for K = 640 is converted to numerical data in the lookup table of FIG. As described above with respect to FIG. 6 and using sliding window averaging, after calculating the average number of iterative processes used to generate the corrected data sequence, the estimation of the frame error rate is, for example, linear interpolation over the average number of actual iterative processes. Can be obtained from the lookup table of FIG.

The comparison simulation results will be discussed with reference to FIGS. 11, 12, 13, 14, 15, 16, 20, 21, and 22 for receivers using conventional error estimation techniques. In particular, conventional receivers generally calculate the frame error rate by monitoring a predetermined number of decoded data frames to provide a frame error rate, counting the number of data frames in the error, and dividing the error number by a predetermined number of decoded data frames. . As shown, conventional error estimation can be made more reliable at receivers experiencing a constant frame error rate by monitoring a large predetermined number of data frames. However, this estimation cannot be sufficiently adapted to the condition that the frame error rate, which is relatively suitable for the frame period, changes.

In Figures 9-18, simulation results are provided to a receiver that depends on a constant signal-to-noise ratio (SNR) for the duration of the 10,000 frames received. In particular, this simulation uses constant operation E _b / N ₀ simulated at 0.9 dB, 1.2 dB, and 1.48 dB with corresponding frame error rates of 0.1014, 0.0111, and 0.0013, respectively. 9, 11, 13, 15, and 17 are provided on a log-log scale to illustrate the initial convergence rate. 10, 12, 14, 16, and 18-22 are provided on a semilog scale to illustrate long term changes.

In Figures 19-22, the simulation results are provided to a receiver that depends on the signal-to-noise ratio that changes over the duration of the 10,000 frames received. The table in FIG. 8 identifies the simulated behavior E _b / N ₀ and the behavior frame error rate for each group of frames within 10,000 frames of the simulation. In particular, a frame error rate of 1% is simulated for frames 1-2000, 4001-6000, and 8001-10000; A frame error rate of 0.1% is simulated for 2001-4000; A frame error rate of 10% is simulated for 6001-8000.

9 and 10 show a WCDMA turbo code receiver (log-max decoder) according to an embodiment of the present invention, which is performed with three operations E _b / N ₀ and a frame error rate, and is performed with a sliding window averaging size (L) of 100 frames. And frame size (K = 640) and sliding window averaging). As shown in Figure 9, a reliable estimate of the frame error rate can be obtained with all three error rates within nearly 30 frames. In addition, the long-term estimation of the frame error rate is reliable as shown in FIG. In particular, the long time estimation of the frame error rate at 10% and 1% is almost centered on the actual frame error rate. The estimate for the actual frame error rate of 0.1% is centered slightly lower than the actual frame error rate. This may be the result of a relatively low sampling resolution in the table of FIG. 7 consisting of steep sloped portions of the graph of FIG. 4B.

11 and 12 show simulation results for a conventional receiver corresponding to the results of FIGS. 9 and 10 and using a window size of 100 frames. That is, the estimation of the frame error rate is generally calculated by monitoring the most recent 100 frames, by counting the number of frame errors, and by dividing the number of frame errors by the number of monitored frames. As shown, no reliable estimate of the frame error rate is provided until at least 1 / (actual FER) frame is monitored. As shown, a reliable estimate of the frame error rate is obtained for FER = 10% after nearly 20 frames. Estimation of the frame error rate for FER = 1% is very unreliable and can sometimes only provide an estimate because it is a number 100 frame window with no frame error resulting from an estimated frame error rate of zero for this estimate. At 0.1% actual FER, the simulation of conventional error estimation does not appear to work in this simulation.

The window size for conventional error estimation of a conventional receiver is increased to 500 frames in the simulations shown in FIGS. 13 and 14. With this increase in window size, conventional receivers, including conventional error estimates, can improve long time estimation of frame errors for actual FER = 1%. However, the estimate at actual FER = 1% using the 500 frame window size is much less reliable than the estimate provided according to the present invention using a smaller window size of 100 frames as shown in FIG. In addition, conventional error estimation using a window size of 500 frames still cannot provide a reliable estimate of the frame error rate when the actual FER is 0.1%.

The window size for conventional error estimation is further increased to 1000 frames in the simulations shown in FIGS. 15 and 16. A reliable estimate of the frame error rate for the actual FER = 10% and 1% is obtained as shown. However, estimation of frame error rate for actual FER = 0.1% still does not provide continuously reliable results in this simulation.

17 and 18 show a WCDMA turbo code receiver (log-) according to an embodiment of the present invention performed with three operations E _b / N ₀ and a frame error rate, and with an increased sliding window averaging size (L) of 1000 frames. Simulation results for the maximum decoder and frame size (K = 640) and sliding window averaging are shown. As shown, a more reliable long time estimate of the frame error rate can be provided when compared to a simulation according to an embodiment of the present invention using a sliding window size L of 100 frames as shown in FIGS. 9 and 10. Can be. As discussed above with respect to FIGS. 9 and 10, a consistent underestimation of the frame error rate is monitored when the actual FER is 0.01, and this underestimation may result from a lack of sufficient sampling resolution in this range of the interpolation table of FIG. 7. Can be. Underestimation can be reduced by providing a higher resolution sample to this portion of the interpolation table of FIG. 7 with a relatively low frame error rate and a relatively low number of iterations used to process received symbol sequences.

19 is a WCDMA turbo code receiver (log-maximum decoder and frame size) according to an embodiment of the present invention performed with varying operation E _b / N ₀ and frame error rate, with a sliding window averaging size L of 100 frames. (K = 640) and sliding window averaging) are shown. In particular, frames 1-2000, 4001-6000, and 8001-10000 are received with actual FER = 1%; Frames 2001-4000 are received with actual FER = 0.1%; Frames 6001-8000 are received with actual FER = 10%. As shown, an error estimator in accordance with an embodiment of the present invention responds reliably and tracks the operating FER.

20-22 show simulation results for a conventional receiver that is dependent on the same varying operating FER as discussed above with respect to FIG. 19 and uses window sizes of 100, 500, and 1000 frames, respectively. That is, the estimation of the frame error rate is generally calculated by monitoring the most recent 100, 500, and 1000 frames, counting the number of frame errors, and dividing the number of frame errors by the number of monitored frames. As shown, conventional error estimation cannot provide a reliable estimate of the operating frame error rate. At a relatively low window size of 100 frames, the simulation of FIG. 20 does not provide a reliable estimate of the low operating frame error rate, where the actual FER is 1% and 0.1%. At high window sizes of 500 and 1000, the simulations of FIGS. 21 and 22 are not adapted to change rapidly with motion frame error rates.

As discussed above, a receiver in accordance with an embodiment of the present invention may thus generate an estimate of the error rate for received communication using the number of iterative baseband processor iterations performed in the calculation of the data frame. The use of iterative baseband processor iterative processing to produce an estimate of the error rate may provide a reliable estimate that is adapted to quickly change the reception conditions. This error estimate can then be sent to the wireless device transmitting the incoming communication, whereby the remote wireless device changes the transmission characteristics, for example the transmit power, and maintains the desired level of reception quality at the receiver. Alternatively, or in addition, the error estimate may be used by the receiver with or without transmitting this error estimate to another wireless device.

The receiver according to the present invention can be used in a communication device such as the mobile terminal 101 shown in FIG. 23, where the mobile terminal 101 provides communication to a base station. As shown, the mobile terminal may include an antenna 99, a receiver 103, a transmitter 105, a controller 107, and a user interface 109, where the user interface includes a keypad, display, speaker, And / or a microphone. In particular, it may be performed like the receiver of FIGS. 1 and 2, and the information bits and error estimates generated by the receiver 103 may be provided to the controller 107. Thus, the controller 107 can provide the error estimate to the transmitter 105 for transmission back to the base station. Accordingly, the mobile terminal 101 can generate an error estimate for the communication received from the base station and send the error estimate back to the base station.

Mobile terminal 101 may be, for example, a wireless telephone, a personal digital assistant, a computer with a wireless modem, or any other device that provides voice and / or data communication. In addition, the receiver according to the invention can be used, for example, in fixed terminals, communication base stations, satellite receivers, or pagers.

The drawings and specification disclose exemplary exemplary embodiments of the invention and, although specific terms are used, these terms are used only in general and descriptive sense and are not used for limitation, the scope of the invention is defined in the following claims It is.

Claims (21)

A method of estimating an error rate for a received communication, comprising: Iteratively processing each of the plurality of received symbol sequences to provide each of the plurality of data sequences; Providing a number of iterations performed to provide each of the plurality of data sequences; And Estimating an error rate for the received communication based on the number of iterations performed to provide each of the plurality of data sequences, Estimating the error rate comprises calculating an average of the number of iterations performed to provide each of the plurality of data sequences such that the error rate is based on an average of the number of iterations performed. Error rate estimation method for. The method of claim 1, Providing a lookup table comprising a candidate mean of the iterations performed and a corresponding estimated error rate for each of the candidate means, wherein estimating the error rate comprises: estimating the estimated error corresponding to the calculated mean. And selecting from a lookup table. The method of claim 1, Iteratively processing each of the plurality of received symbol sequences includes: Iteratively processing the received symbol sequence to provide an estimated data sequence after each iteration; Performing an error check on the estimated data sequence after each iteration; Performing subsequent iterative processing of the received symbol sequence in response to detecting an error during the error check; And Terminating subsequent iterative processing of the received symbol sequence in response to not detecting an error during error checking and transmitting the estimated data sequence as the data sequence. Way. The method of claim 3, wherein And performing an error check comprises performing a cyclic redundancy check (CRC). The method of claim 1, And the estimated error rate comprises an estimated bit error rate. The method of claim 1, And the data sequence comprises a data frame, and wherein the estimated error rate comprises an estimated frame error rate. In the receiver: Means for receiving a plurality of received symbol sequences; Means for iteratively processing each of the plurality of received symbol sequences to provide each of the plurality of data sequences; Means for providing a number of iterations performed to provide each of the plurality of data sequences; And Means for estimating an error rate for the received communication based on the number of iterations performed to provide each of the plurality of data sequences, And the means for estimating the error rate comprises means for calculating an average of the number of iterations performed to provide each of the plurality of data sequences so that the error rate is based on the average of the number of iterations performed. The method of claim 7, wherein Means for providing a lookup table that includes a candidate mean of the iterations performed and a corresponding estimated error rate for each of the candidate means, wherein the means for estimating the error rate comprises estimating an estimated error corresponding to the calculated mean. And means for selecting from a lookup table. The method of claim 7, wherein Means for iteratively processing each of the plurality of received symbol sequences are: Means for iteratively processing the received symbol sequence to provide an estimated data sequence after each iteration; Means for performing an error check on the estimated data sequence after each iteration; Means for performing subsequent iterative processing of the received symbol sequence in response to detecting an error during the error check; And Means for terminating subsequent iterative processing of the received symbol sequence in response to not detecting an error during error checking and transmitting the estimated data sequence as the data sequence. The method of claim 9, Means for performing the error check comprises means for performing a cyclic redundancy check (CRC). The method of claim 7, wherein And the estimated error rate comprises an estimated bit error rate. The method of claim 7, wherein The data sequence comprises a data frame and the estimated error rate comprises an estimated frame error rate. In the receiver: A radio processor for generating a plurality of received symbol sequences; An iterative baseband processor coupled to the radio processor, repetitively processing each of the plurality of received symbol sequences to provide each of a plurality of data sequences; Detection circuitry coupled to the repeatable baseband processor, the detection circuit providing a number of iterations performed to provide each of the plurality of data sequences; And An error estimator coupled to the detection circuit, for estimating an error rate for the received communication based on the number of iterations performed to provide each of the plurality of data sequences, And the error estimator is further configured to calculate an average of the number of iterations performed to provide each of a plurality of data sequences to provide an error rate based on the average of the number of iterations performed. The method of claim 13, And further comprising a lookup table comprising a candidate mean of each performed iteration and a corresponding estimated error rate for each candidate mean, wherein the error estimator selects from the lookup table an estimated error corresponding to the calculated mean. Receiver. The method of claim 13, The iterative baseband processor is further configured to iteratively process the received symbol sequence to provide an estimated data sequence after each iteration, and the detection circuitry is also to check for errors for the estimated data sequence after each iteration. Wherein the repetitive baseband processor performs subsequent iteration processing of the received symbol sequence in response to detecting an error during the error check, and the repetitive baseband processor does not detect an error during the error check. Responsive to not ending the subsequent iterative processing of the received symbol sequence and transmitting the estimated data sequence as a data sequence. The method of claim 15, And said detecting circuit comprises a cyclic redundancy check (CRC) circuit. The method of claim 13, And the estimated error rate comprises an estimated bit error rate. The method of claim 13, The data sequence comprises a data frame and the estimated error rate comprises an estimated frame error rate. delete delete delete

Images